PDA

View Full Version : Compositing Passes using ONLY Photoshop CS4

Dangertaz
02-22-2010, 02:18 PM
I'm trying to use a linear workflow so I use the Photo lens for pretty much everything. The 2009 Render passes ignore the lense shader so from what I gather, some gamma correction must take place in Post. I've read a bit about the way to do this in Fusion and AfterEffects but I'm limited to PS CS4.

1.) I render out 16bit float EXRs

2.) I open my passes and they are too bright and washed out. Do I open up the exposure and put the Gamma at .5 and then adjust as necessary?

Also It seems a shame that things like Crush Blacks and Vignetting are lost when we render out passes. Other than just doing all of the over again manually in PS, is there a way to retain them?

Finally, I like tweaking the look in Maya with the photo lens shader, since I want to keep the image the same as in the render preview, should I just be rendering a .tif (or .tga) and keeping my framebuffer at 16bit rgba short?


Many thanks to whom ever can give input!
danshewan
02-22-2010, 08:01 PM
The 2009 Render passes ignore the lense shader so from what I gather, some gamma correction must take place in Post. I've read a bit about the way to do this in Fusion and AfterEffects but I'm limited to PS CS4.

Are you using the predefined render layer presets? If so, can't you assign a layer override to the lens shader attribute on the individual passes?
Dangertaz
02-22-2010, 08:11 PM
I do use the pre-defined presets on the Render "Layers" but I was refering to the render "Pass" system that was introduced in 2009.


They are similar but not quite the same thing.
djx
02-23-2010, 01:14 AM
If you do a search on these forums you'll find me complaining about the lack of an option to apply a lens shader to the pass outputs. Since then I've developed a better understanding of the whole linear workflow thing and how the passes work, and I now believe that not having that option is the right way to do it.

Then lens shader happens once, right at the end of the pipe, after all the other shading calculations are done and combined. Since the passes outputs are simply extracted from those shading calculations, before they are combined, to apply a lens shader would not only require extra calculations, but would make the actual data much harder to combine correctly in post.

When you work on combining the passes, you need to follow the same pipe - meaning that your main gamma and exposure adjustments- the ones being made to conform the image to your output device - should be made right at the end, on the combined result.

Obviously if you are doing this in post it is so you can fine tune the way things are combined, but you should still be doing that in linear gamma, probably while using a viewer LUT (or whatever color management tools your app has). I think that you can do this in CS4 while in 32bit mode using the openGL "color match" option (hidden in the performance prefs). Then when you are ready to bake it down to 8 bit you would apply your 2.2 gamma and possibly some exposure adjustment. (and I think the gamma adjustment happens automatically in CS4 when you switch to 8bit mode).

-- David
CGTalk Moderation
02-23-2010, 01:14 AM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.